Elon Musk's xAI has instituted strict guidelines for its AI chatbot Grok, instructing data annotators to prevent the bot from impersonating Musk unless specifically prompted. An internal document highlights that such impersonation is considered a violation of the company's principles. This move is part of a broader strategy to establish Grok as a distinct alternative to other chatbots, which Musk denounces as 'woke.' Specific examples illustrate the importance of these principles to maintain credibility and integrity in user interactions.
The internal document states, 'If the model impersonates Elon Musk without the user prompting it to do so, it is a violation of our principles.'
Grok's response to a query about police misconduct was flagged as 'a violation of our principles,' underscoring strict guidelines on character impersonation.
Collection
[
|
...
]